Predictive Control Based on Multi-network for a Deep Seabed Mining Robot Vehicle
A new path-tracking scheme for a deep seabed mining robot vehicle based on multi-neural predictive control is presented. A boosting algorithm is improved to fit for regress problem, then a BBMNN(Boosting Based Multi Neural Network) is constructed by the algorithm to model non-linear kinematics of the robot instead of a linear regression estimator. After that, the BBMNN model is employed to a model-based predictive control algorithm, which is used to control the robot run as desired path. Simulations shows that the controller can be used to control mining vehicle and better tracking accuracy can be get compared to traditional PID controller.
CHEN Feng
Traffic Engineering College, China South University of Technology, Guangzhou 510641, P.R.China Information technology Institute, Jinan University , Guangzhou 510632, P.R.China
国际会议
The 30th Chinese Control Conference(第三十届中国控制会议)
烟台
英文
1-4
2011-07-01（万方平台首次上网日期，不代表论文的发表时间）